1
|
Wang T, Zhang M, Zhang J, Ng WWY, Chen CLP. BASS: Broad Network Based on Localized Stochastic Sensitivity. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024; 35:1681-1695. [PMID: 35830397 DOI: 10.1109/tnnls.2022.3184846] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/15/2023]
Abstract
The training of the standard broad learning system (BLS) concerns the optimization of its output weights via the minimization of both training mean square error (MSE) and a penalty term. However, it degrades the generalization capability and robustness of BLS when facing complex and noisy environments, especially when small perturbations or noise appear in input data. Therefore, this work proposes a broad network based on localized stochastic sensitivity (BASS) algorithm to tackle the issue of noise or input perturbations from a local perturbation perspective. The localized stochastic sensitivity (LSS) prompts an increase in the network's noise robustness by considering unseen samples located within a Q -neighborhood of training samples, which enhances the generalization capability of BASS with respect to noisy and perturbed data. Then, three incremental learning algorithms are derived to update BASS quickly when new samples arrive or the network is deemed to be expanded, without retraining the entire model. Due to the inherent superiorities of the LSS, extensive experimental results on 13 benchmark datasets show that BASS yields better accuracies on various regression and classification problems. For instance, BASS uses fewer parameters (12.6 million) to yield 1% higher Top-1 accuracy in comparison to AlexNet (60 million) on the large-scale ImageNet (ILSVRC2012) dataset.
Collapse
|
2
|
Zhang X, Zhong C, Zhang J, Wang T, Ng WW. Robust Recurrent Neural Networks for Time Series Forecasting. Neurocomputing 2023. [DOI: 10.1016/j.neucom.2023.01.037] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/15/2023]
|
3
|
Hassan E, Shams MY, Hikal NA, Elmougy S. The effect of choosing optimizer algorithms to improve computer vision tasks: a comparative study. MULTIMEDIA TOOLS AND APPLICATIONS 2022; 82:16591-16633. [PMID: 36185324 PMCID: PMC9514986 DOI: 10.1007/s11042-022-13820-0] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 01/30/2022] [Revised: 06/30/2022] [Accepted: 09/06/2022] [Indexed: 06/16/2023]
Abstract
Optimization algorithms are used to improve model accuracy. The optimization process undergoes multiple cycles until convergence. A variety of optimization strategies have been developed to overcome the obstacles involved in the learning process. Some of these strategies have been considered in this study to learn more about their complexities. It is crucial to analyse and summarise optimization techniques methodically from a machine learning standpoint since this can provide direction for future work in both machine learning and optimization. The approaches under consideration include the Stochastic Gradient Descent (SGD), Stochastic Optimization Descent with Momentum, Rung Kutta, Adaptive Learning Rate, Root Mean Square Propagation, Adaptive Moment Estimation, Deep Ensembles, Feedback Alignment, Direct Feedback Alignment, Adfactor, AMSGrad, and Gravity. prove the ability of each optimizer applied to machine learning models. Firstly, tests on a skin cancer using the ISIC standard dataset for skin cancer detection were applied using three common optimizers (Adaptive Moment, SGD, and Root Mean Square Propagation) to explore the effect of the algorithms on the skin images. The optimal training results from the analysis indicate that the performance values are enhanced using the Adam optimizer, which achieved 97.30% accuracy. The second dataset is COVIDx CT images, and the results achieved are 99.07% accuracy based on the Adam optimizer. The result indicated that the utilisation of optimizers such as SGD and Adam improved the accuracy in training, testing, and validation stages.
Collapse
Affiliation(s)
- Esraa Hassan
- Faculty of Artificial Intelligence, Kafrelsheikh University, Kafrelsheikh, 33516 Egypt
| | - Mahmoud Y. Shams
- Faculty of Artificial Intelligence, Kafrelsheikh University, Kafrelsheikh, 33516 Egypt
| | - Noha A. Hikal
- Department of Information Technology, Faculty of Computers and Information, Mansoura University, Mansoura, 35516 Egypt
| | - Samir Elmougy
- Department of Computer Science, Faculty of Computers and Information, Mansoura University, Mansoura, 35516 Egypt
| |
Collapse
|
4
|
|
5
|
Bilateral sensitivity analysis: a better understanding of a neural network. INT J MACH LEARN CYB 2022. [DOI: 10.1007/s13042-022-01511-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/19/2022]
|
6
|
An automatic framework for perioperative risks classification from retinal images of complex congenital heart disease patients. INT J MACH LEARN CYB 2021. [DOI: 10.1007/s13042-021-01419-0] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/23/2022]
|
7
|
Ng WW, Liu Z, Zhang J, Pedrycz W. Maximizing minority accuracy for imbalanced pattern classification problems using cost-sensitive Localized Generalization Error Model. Appl Soft Comput 2021. [DOI: 10.1016/j.asoc.2021.107178] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/22/2022]
|
8
|
A stochastic sensitivity-based multi-objective optimization method for short-term wind speed interval prediction. INT J MACH LEARN CYB 2021. [DOI: 10.1007/s13042-021-01340-6] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/21/2022]
|
9
|
Wang T, Ng WWY, Pelillo M, Kwong S. LiSSA: Localized Stochastic Sensitive Autoencoders. IEEE TRANSACTIONS ON CYBERNETICS 2021; 51:2748-2760. [PMID: 31331899 DOI: 10.1109/tcyb.2019.2923756] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/10/2023]
Abstract
The training of autoencoder (AE) focuses on the selection of connection weights via a minimization of both the training error and a regularized term. However, the ultimate goal of AE training is to autoencode future unseen samples correctly (i.e., good generalization). Minimizing the training error with different regularized terms only indirectly minimizes the generalization error. Moreover, the trained model may not be robust to small perturbations of inputs which may lead to a poor generalization capability. In this paper, we propose a localized stochastic sensitive AE (LiSSA) to enhance the robustness of AE with respect to input perturbations. With the local stochastic sensitivity regularization, LiSSA reduces sensitivity to unseen samples with small differences (perturbations) from training samples. Meanwhile, LiSSA preserves the local connectivity from the original input space to the representation space that learns a more robustness features (intermediate representation) for unseen samples. The classifier using these learned features yields a better generalization capability. Extensive experimental results on 36 benchmarking datasets indicate that LiSSA outperforms several classical and recent AE training methods significantly on classification tasks.
Collapse
|
10
|
Hamadneh NN. Dead Sea Water Levels Analysis Using Artificial Neural Networks and Firefly Algorithm. INTERNATIONAL JOURNAL OF SWARM INTELLIGENCE RESEARCH 2020. [DOI: 10.4018/ijsir.2020070102] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/09/2022]
Abstract
In this study, the performance of adaptive multilayer perceptron neural network (MLPNN) for predicting the Dead Sea water level is discussed. Firefly Algorithm (FFA), as an optimization algorithm is used for training the neural networks. To propose the MLPNN-FFA model, Dead Sea water levels over the period 1810–2005 are applied to train MLPNN. Statistical tests evaluate the accuracy of the hybrid MLPNN-FFA model. The predicted values of the proposed model were compared with the results obtained by another method. The results reveal that the artificial neural network (ANN) models exhibit high accuracy and reliability for the prediction of the Dead Sea water levels. The results also reveal that the Dead Sea water level would be around -450 until 2050.
Collapse
|
11
|
|
12
|
Ng WW, Xu S, Wang T, Zhang S, Nugent C. Radial Basis Function Neural Network with Localized Stochastic-Sensitive Autoencoder for Home-Based Activity Recognition. SENSORS 2020; 20:s20051479. [PMID: 32182668 PMCID: PMC7085686 DOI: 10.3390/s20051479] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 01/27/2020] [Revised: 02/28/2020] [Accepted: 03/03/2020] [Indexed: 11/16/2022]
Abstract
Over the past few years, the Internet of Things (IoT) has been greatly developed with one instance being smart home devices gradually entering into people's lives. To maximize the impact of such deployments, home-based activity recognition is required to initially recognize behaviors within smart home environments and to use this information to provide better health and social care services. Activity recognition has the ability to recognize people's activities from the information about their interaction with the environment collected by sensors embedded within the home. In this paper, binary data collected by anonymous binary sensors such as pressure sensors, contact sensors, passive infrared sensors etc. are used to recognize activities. A radial basis function neural network (RBFNN) with localized stochastic-sensitive autoencoder (LiSSA) method is proposed for the purposes of home-based activity recognition. An autoencoder (AE) is introduced to extract useful features from the binary sensor data by converting binary inputs into continuous inputs to extract increased levels of hidden information. The generalization capability of the proposed method is enhanced by minimizing both the training error and the stochastic sensitivity measure in an attempt to improve the ability of the classifier to tolerate uncertainties in the sensor data. Four binary home-based activity recognition datasets including OrdonezA, OrdonezB, Ulster, and activities of daily living data from van Kasteren (vanKasterenADL) are used to evaluate the effectiveness of the proposed method. Compared with well-known benchmarking approaches including support vector machine (SVM), multilayer perceptron neural network (MLPNN), random forest and an RBFNN-based method, the proposed method yielded the best performance with 98.35%, 86.26%, 96.31%, 92.31% accuracy on four datasets, respectively.
Collapse
Affiliation(s)
- Wing W.Y. Ng
- Guangdong Provincial Key Lab of Computational Intelligence and Cyberspace Information, School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China; (W.W.Y.N.); (S.X.)
| | - Shichao Xu
- Guangdong Provincial Key Lab of Computational Intelligence and Cyberspace Information, School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China; (W.W.Y.N.); (S.X.)
| | - Ting Wang
- Guangdong Provincial Key Lab of Computational Intelligence and Cyberspace Information, School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China; (W.W.Y.N.); (S.X.)
- Correspondence: ; Tel.: +86-132-8868-8360
| | - Shuai Zhang
- School of Computing, Ulster University, Shore Road, Newtownabbey, Co., Antrim BT37 0QB, Northern Ireland, UK; (S.Z.); (C.N.)
| | - Chris Nugent
- School of Computing, Ulster University, Shore Road, Newtownabbey, Co., Antrim BT37 0QB, Northern Ireland, UK; (S.Z.); (C.N.)
| |
Collapse
|
13
|
Fan H, Pei J, Zhao Y. An optimized probabilistic neural network with unit hyperspherical crown mapping and adaptive kernel coverage. Neurocomputing 2020. [DOI: 10.1016/j.neucom.2019.09.029] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/26/2022]
|
14
|
Modeling and Optimization of Gaseous Thermal Slip Flow in Rectangular Microducts Using a Particle Swarm Optimization Algorithm. Symmetry (Basel) 2019. [DOI: 10.3390/sym11040488] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022] Open
Abstract
In this study, pressure-driven flow in the slip regime is investigated in rectangular microducts. In this regime, the Knudsen number lies between 0.001 and 0.1. The duct aspect ratio is taken as 0 ≤ ε ≤ 1 . Rarefaction effects are introduced through the boundary conditions. The dimensionless governing equations are solved numerically using MAPLE and MATLAB is used for artificial neural network modeling. Using a MAPLE numerical solution, the shear stress and heat transfer rate are obtained. The numerical solution can be validated for the special cases when there is no slip (continuum flow), ε = 0 (parallel plates) and ε = 1 (square microducts). An artificial neural network is used to develop separate models for the shear stress and heat transfer rate. Both physical quantities are optimized using a particle swarm optimization algorithm. Using these results, the optimum values of both physical quantities are obtained in the slip regime. It is shown that the optimal values ensue for the square microducts at the beginning of the slip regime.
Collapse
|
15
|
A robust correlation analysis framework for imbalanced and dichotomous data with uncertainty. Inf Sci (N Y) 2019. [DOI: 10.1016/j.ins.2018.08.017] [Citation(s) in RCA: 41] [Impact Index Per Article: 8.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
|
16
|
Kowalski PA, Kusy M. Sensitivity Analysis for Probabilistic Neural Network Structure Reduction. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2018; 29:1919-1932. [PMID: 28422668 DOI: 10.1109/tnnls.2017.2688482] [Citation(s) in RCA: 16] [Impact Index Per Article: 2.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/07/2023]
Abstract
In this paper, we propose the use of local sensitivity analysis (LSA) for the structure simplification of the probabilistic neural network (PNN). Three algorithms are introduced. The first algorithm applies LSA to the PNN input layer reduction by selecting significant features of input patterns. The second algorithm utilizes LSA to remove redundant pattern neurons of the network. The third algorithm combines the proposed two and constitutes the solution of how they can work together. PNN with a product kernel estimator is used, where each multiplicand computes a one-dimensional Cauchy function. Therefore, the smoothing parameter is separately calculated for each dimension by means of the plug-in method. The classification qualities of the reduced and full structure PNN are compared. Furthermore, we evaluate the performance of PNN, for which global sensitivity analysis (GSA) and the common reduction methods are applied, both in the input layer and the pattern layer. The models are tested on the classification problems of eight repository data sets. A 10-fold cross validation procedure is used to determine the prediction ability of the networks. Based on the obtained results, it is shown that the LSA can be used as an alternative PNN reduction approach.
Collapse
|
17
|
Prediction of thermal conductivity of polyvinylpyrrolidone (PVP) electrospun nanocomposite fibers using artificial neural network and prey-predator algorithm. PLoS One 2017; 12:e0183920. [PMID: 28934220 PMCID: PMC5608192 DOI: 10.1371/journal.pone.0183920] [Citation(s) in RCA: 16] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/26/2017] [Accepted: 08/08/2017] [Indexed: 11/19/2022] Open
Abstract
In this study, multilayer perception neural network (MLPNN) was employed to predict thermal conductivity of PVP electrospun nanocomposite fibers with multiwalled carbon nanotubes (MWCNTs) and Nickel Zinc ferrites [(Ni0.6Zn0.4) Fe2O4]. This is the second attempt on the application of MLPNN with prey predator algorithm for the prediction of thermal conductivity of PVP electrospun nanocomposite fibers. The prey predator algorithm was used to train the neural networks to find the best models. The best models have the minimal of sum squared error between the experimental testing data and the corresponding models results. The minimal error was found to be 0.0028 for MWCNTs model and 0.00199 for Ni-Zn ferrites model. The predicted artificial neural networks (ANNs) responses were analyzed statistically using z-test, correlation coefficient, and the error functions for both inclusions. The predicted ANN responses for PVP electrospun nanocomposite fibers were compared with the experimental data and were found in good agreement.
Collapse
|
18
|
|
19
|
Ng WWY, Hu J, Yeung DS, Yin S, Roli F. Diversified Sensitivity-Based Undersampling for Imbalance Classification Problems. IEEE TRANSACTIONS ON CYBERNETICS 2015; 45:2402-2412. [PMID: 25474818 DOI: 10.1109/tcyb.2014.2372060] [Citation(s) in RCA: 66] [Impact Index Per Article: 7.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/04/2023]
Abstract
Undersampling is a widely adopted method to deal with imbalance pattern classification problems. Current methods mainly depend on either random resampling on the majority class or resampling at the decision boundary. Random-based undersampling fails to take into consideration informative samples in the data while resampling at the decision boundary is sensitive to class overlapping. Both techniques ignore the distribution information of the training dataset. In this paper, we propose a diversified sensitivity-based undersampling method. Samples of the majority class are clustered to capture the distribution information and enhance the diversity of the resampling. A stochastic sensitivity measure is applied to select samples from both clusters of the majority class and the minority class. By iteratively clustering and sampling, a balanced set of samples yielding high classifier sensitivity is selected. The proposed method yields a good generalization capability for 14 UCI datasets.
Collapse
|